st

The 1 International Conference on Water Resources and Coastal Engineering

HYDROLOGIC ALTERATION UNDER ANTHROPOGENIC AND CLIMATE

CHANGES IN VU GIA – THU BON

T.T.Vu1,2, J.Kiesel3,1, B.Guse1,4, N.Fohrer1

1
Department of Hydrology and Water Resources Management, Institute for Natural Resource Conservation, Kiel,
Germany; E-mail: tinhvu@hydrology.uni-kiel.de; nfohrer@hydrology.uni-kiel.de
2
Faculty of Water Resource Engineering, The University of Danang University of Science and Technology, Danang,
Vietnam
3
Department of Ecosystem Research, Leibniz-Institute of Freshwater Ecology and Inland Fisheries (IGB), Berlin,
Germany; E-mail: kiesel@igb-berlin.de
4
GFZ German Research Centre for Geosciences, Section Hydrology, Potsdam, Germany
E-mail: bfguse@gfz-potsdam.de

Abstract - The Vu Gia - Thu Bon catchment is the impound the rivers, and to build a system of hydropower
largest inland river basin in central Vietnam and was reservoir cascades in certain cases, become one of the
subject to both anthropogenic and climatic changes during most considerable anthropogenic disturbances of natural
the last decades. In this study, we improve the cycles of water in basins (Costigan and Daniels, 2012).
understanding of the hydrologic alteration in this area As one of the largest inland river basin in central
caused by changes in precipitation pattern and damming. Vietnam, the Vu Gia-Thu Bon (VGTB) catchment has
Therefore, we split the discharge time series (1976-2015) been subject to both anthropogenic and climatic changes
in Thanh My and Nong Son at the abrupt change in (1) during the last decades. According to Souvignet et al.,
precipitation records in 1995 and (2) the year of operating (2014) rainfall in VGTB shows an increasing trend since
the first dam in the two catchments. A set of 32 the 1980s which explains up to 74% of change in
hydrologic indicators was used to evaluate the discharge. However, the inaccuracy of the overall trend in
hydrological alteration between the pre- and post-period. long time series can be a source of inaccuracy if there is a
Results show a similar impact pattern resulting from sudden change at some point during the recorded period.
climatic factors, with an increase in river flow. The (Rahmani et al., 2015). By using a combination of
anthropogenically related factors caused contrasting statistical tests, Vu et al., (2018) have found an abrupt
results in the two neighboring catchments to a greater change in 1995 occurring in the rainfall data upstream of
extent than by climate, especially for low flow conditions. VGTB. This shift therefore should be taken into
The diversion dams could be responsible for a decreased consideration to investigate the effect of this climatic
monthly flow and low flow during dry season in Thanh factor in the river regime.
My, accounting for 55% to 70% of reduction in discharge. Due to the topography and high discharges, between
This quantitative comparison is essential, not only for 2005 to 2015, 24 hydropower plants were newly built in
understanding the hydrological response in the catchment the whole catchment, and are able to store approximately
but also for local water resources management. 2.16 billion m3 of water. Alterations in the flow regime
Keywords – Vu Gia-Thu Bon; IHA; trend; are the major consequences and need to be well studied.
anthropogenic impact; climate change Therefore, this area provides an excellent case for an
investigation of comparing changes due to climate and
human factors.
1. Introduction
Though over 60% of catchment area is covered by
Climatic and anthropogenic changes have been widely
natural mountainous areas and forests, Laux et al. (2016)
proven to be responsible to the change in hydrologic
pointed out that deforestation in the area caused no
regime, and therefore have raised the interest in
observable impacts to surface air temperature, and
catchment hydrology over the last decades. Under global
negligible impacts to the precipitation as well as the
climate change and local anthropogenic alterations,
discharge regime. Thus, attempts to investigate the impact
hydrological regime may suffer significant modification.
of land use changes on the discharge regime were not
Climate impacts, particularly in precipitation patterns, made in this paper.
are likely to enhance the frequency and severity of
Effects of climatic- and anthropogenic changes on
extreme climate events (Ye et al., 2013). It implies that
runoff variation are significantly sensitive and resulted in
gradual changes in climate factors might have little effect
serve environmental degradation. To characterize
until a threshold is reached at which a large shift occurs
statistical properties of the long-term hydrologic regime
that might be difficult to reverse (Scheffer and Carpenter,
of rivers, hydrologic indicators which are grouped into
2003). Therefore, investigating a certain shift in advance
five fundamental properties (Richter et al., 1996) are often
may help to better understand the change in hydrology.
used. A number of studies using this set of indicators is
Unlike the climate factor, where the changes are more listed in Vu et al., (2018).
likely supposed to be gradual, the existing dams that
 T.T.Vu, J.Kiesel, B.Guse, N.Fohrer 2
Several authors have investigated climate changes in 2.2. Methodology
the river regimes in VGTB (Do et al., 2012; Souvignet et 2.2.1. Trend detection
al., 2014; Laux et al., 2016). To our knowledge, the
To analyze significant changes in mean flow and
relative contribution of climate change and human
runoff extremes, the non-parametric Mann-Kendall (MK)
activities to runoff change in the Vu Gia Thu Bon
trend test was used. The test is rank-based and therefore
catchment however has not been well investigated. In this
the assumption of normality in the distribution is not
study, we aim to investigate the hydrologic alteration
required. The null hypothesis of “no-trend” H0 and the
under changes in precipitation patterns and dam
alternative hypothesis of “monotonic trend” H1 are
occurrence and to answer the question to which extent
considered with a pre-assigned significance level
these factors might have altered the hydrologic regime.
(p.value) of 5%. The presence of a trend is evaluated
2. Methodology using the Z value. A positive (negative) value of Z
2.1. Study area indicates a positive (negative) trend. In this study, MK
The Vu Gia-Thu Bon catchment is formed by two test was used to detect changes in six meteorological time
main rivers named Vu Gia and Thu Bon which are series in six available rainfall gauges in study area by
connected by the natural channel Quang Hue in the using „trend‟ R-package (Pohlert, 2016).
downstream, and the reservoir cascade Dak Mi 4 in the 2.2.2. Hydrologic alteration measurement
upstream (Fig. 1). The climate in this catchment is wet- The hydrologic alteration of the discharge regime
dry tropical with a strong seasonality. Very high rainfall before and after the changes by PREC and DAM were
occurs during the rainy season (September – December) evaluated by using the set of 32 Indicators of Hydrologic
while dry season lasts for the remaining months, with the Alteration (IHA, Table 1) suggested by Richter et al.
lowest flows occurring in August. The diversion of Dak (1996). The indicators were classified in 5 categories:
Mi 4A plant from Vu Gia to Thu Bon (through Dak Mi Magnitude of mean monthly flow (MMF) – Magnitude
4B, C, see Fig. 1) further highlights the anthropogenic and duration of extreme flow (MDE) – Frequency (FRE)
impacts in VGTB. – Timing (TIM) – Rate of change (RATE).
Table 1: Summary of 32 IHA defined by Richter et al. (1996)
Group Description Unit Abbr.
MMF Mean monthly flow from January to December m³/s Jan-Dec
Annual maxima of 1-day mean of discharge m³/s 1dmax
Annual maxima of 3-day mean of discharge m³/s 3dmax
Annual maxima of 7-day mean of discharge m³/s 7dmax
Annual maxima of 30-day mean of discharge m³/s 30dmax
Annual maxima of 90-day mean of discharge m³/s 90dmax
MDE
Annual minima of 1-day mean of discharge m³/s 1dmin
Annual minima of 3-day mean of discharge m³/s 3dmin
Annual minima of 7-day mean of discharge m³/s 7dmin
Annual minima of 30-day mean of discharge m³/s 30dmin
Annual minima of 90-day mean of discharge m³/s 90dmin
High flow pulse duration* days Highpulse
Low flow pulse duration* days Lowpulse
FRE
Fig. 1: The upper part of Vu Gia – Thu Bon catchment Number of high pulses every year -** Highfreq
and the locations of meteo-hydrological gauges Number of low pulses every year - Lowfreq

Change in precipitation is considered as the major Julian date of the maximum day in year - Datemax
TIM
climate related change and henceforth referred to as Julian date of the minimum day in year - Datemin
PREC. To make the change in precipitation more Mean rate of all positive changes in flow m³/s Riserate
understandable, we further tested the trend in the Mean rate of all negative changes in flow m³/s Fallrate
RATE
historical rainfall records of six rainfall gauges (Fig. 1). Number of rises - Nodayrise
The main anthropogenic factor in this study area is Number of falls - Nodayfall
damming (hereafter referred as DAM) and is evaluated by th
*low- and high flows are flow events which below the 25 percentile and above
comparing the time period of pristine condition (no dam) the 75th percentile of the entire flow record; ** these indicators are dimensionless
and the impoundment period. Thus, the year of operating We considered change (C) of indicators (I) in
the first dam in each sub-basin was considered as an percentage for pre- and post-periods as:
abrupt change, which is 2012 (Dak Mi 4) in Thanh My
and 2007 (Khe Dien) in Nong Son. Therefore, daily (1)
discharge time series of the gauges Thanh My and Nong
Son, with a drainage area of 1.850 km2 and 3.155 km² The “EflowStats” R-package (Henriksen et al., 2006)
respectively, were collected from 1976-2015. was used to calculate the hydrologic indicators for the
 st
The 1 International Conference on Water Resources and Coastal Engineering
pre- and post-periods from daily discharge. To interpret In Thanh My, the magnitude of mean flow (MMF)
the changes due to DAM and PREC, we considered (1) and magnitude of extreme flow (MDE) show dominant
the mean changes in the above-defined groups and (2) the changes by DAM. The examined changes are up to two
alteration in single indicator within each group. times higher than those changes related to PREC. The
3. Results average impact (AVE) by DAM is quantified relatively
higher than PREC. This is however in contrast with
3.1.1. Trend in precipitation
changes in Nong Son, where PREC results in more
Table 2 shows the trend in precipitation in a period of significant changes than DAM.
40 years, 1976-2015. There, the positive value of Z
indicates that a slight increasing trend exists in both sub-
basins.
Table 2: Statistical value of trend occurring in six
rainfall gauges
Sub-basins Stations Z p.value
Kham Duc 2.202 0.03*
Vu Gia
Thanh My 1.065 0.29
Tien Phuoc 1.258 0.21
Fig. 3: Alteration of IHA groups (x-axis) in Thanh My and
Tra My 1.427 0.15 Nong Son under two factors (y-axis)
Thu Bon
Hiep Duc 2.129 0.02*
Comparing both catchments, MMF and MDE are
Nong Son 1.089 0.28 subject to the most dominant change while timing is less
*) these indicate the monotonic trend at significance level of 5% impacted in general.
Figure 2 further shows that, fluctuated trends are 3.1.3. Quantifying alteration in hydrologic indicators
observed in most gauges. In Vu Gia sub-basin, a
The changes in precipitation lead to an increase in the
significant trend is observed in Kham Duc while it is not
majority of indicators, while damming results in different
significant in Thanh My, though a positive value is
impacts in both catchments (Fig. 4).
presented. Similarly, rainfall in Hiep Duc in Thu Bon sub-
basin is subject to a significant increase during the period.
A slightly upwards trend is detected in Tra My. Rainfall
in Tien Phuoc and Nong Son show an increasing trend in
the first 30 years before decreasing afterwards.

Fig. 2: Trend in precipitation (in mm, y-axis) in six rainfall

gauges from 1976-2015 (x-axis) Fig. 4: Changes of 32 indicators (y-axis) due to changes
caused by DAM and PREC in Thanh My and Nong Son. Orange
3.1.2. Dominant group of change in pre- and post-period cells present a decrease, while cyan an increase.
Figure 3 shows the average change quantified by Under changes related to PREC, mean monthly flow
PREC and DAM. There, DAM alters the hydrologic increases significantly from 30%-70% in both sub-basins,
indicators more significantly in Thanh My than in Nong with the highest amount of changes in September (about
Son, while PREC has less impact.
 T.T.Vu, J.Kiesel, B.Guse, N.Fohrer 4
70%). This is similar to extreme low and high flow, potentially related to the design of the reservoirs in the
though extreme low flow is subject to increase in a upstream. The Dak Mi 4A (2012) has impounded about
smaller extent as compared to high flow. The similarity is 310 million m3, equivalent to 8% of annual volume in Vu
further shown in Julian date of minimum flow (Datemin) Gia up to Thanh My. The deficit of water in Thanh My
which is subject to occur later in both areas. The after diverting water to Thu Bon is then unavoidable. In
dissimilarity is in frequency of flow events (FRE) which the meanwhile, Khe Dien has stored less than 0.5% of
behave differently in the two catchments. This is therefore annual volume in Thu Bon up to Nong Son. Thus, it is not
not related to precipitation. There, less low flow pulse the main reason of the increase of flow in Nong Son. This
(Lowfreq) occurs in both sub-basins, but with higher is in line with Gao et al. (2009) that storage ratio does not
number of Lowpulse in Thanh My and less in Nong Son. necessarily play a significant role in determining which
Damming results in contradictory changes in the two indicators are most representative of flow regimes. The
catchments. The very significant reduction is observed in operating regulation however should be responsible for
Thanh My, while an increase is highlighted in Nong Son. that (Yang et al., 2008). This diversion further makes the
This conflict stresses in mean flow from March and ends intensity of alteration in the diverted branch more severe
in September, and in extreme low flow. Both high and than the received branch.
low pulses decrease after dam, clearly shows that the This transfer apparently results in changes of
alteration in these indicators is linked to the occurrence of magnitude, duration and timing of extreme flow and rate
dam. A slight decrease in Julian date of minimum flow of changes in the discharge regime. Ordinary hydropower
occurs in Thanh My while this is absent in Nong Son. dam in Nong Son alleviate droughts, with higher flow and
In general, the IHA assessment illustrates a similar less pulse which is below threshold. However, drought
impact pattern resulting from climate related factors in the events under diversion dam are aggravated in the natural
study area, with an increase in river flow. Meanwhile, the river reaches downstream. Evidently, monthly flow in the
anthropogenic factor caused contrasting results in the two early dry season in Thanh My significantly decreases due
neighboring catchments. The diversion dams are likely to the presence of Dak Mi 4. The reduction of flow during
responsible for a decreased monthly flow and low flow the drought period makes the earlier occurrence of
during dry season in Thanh My, accounting for 55% to extreme low flow in Thanh My more explainable.
70% of reduction in discharge. The alteration quantified by dam in both areas has
4. Discussion and conclusion agreed that the flow fluctuates around smaller variation,
where increasing days is expected to occurs more frequent
The estimation of hydrologic alteration in this study
but in lower rate. This means that flow is divided into
has addressed the question to which extent the impacts of
distinct phases and strongly dependent on the regulation
climatic- and anthropogenic have altered the hydrologic
of reservoirs.
regime. We will here discuss the impacts of each
individual factor. In this study, the impact of anthropogenic factors on
the river regime was found to exceed climate-related
Hydrologic alteration caused by climatic changes
factors. This is affirmed by the higher changes in
The precipitation pattern in both sub-basins shows an hydrologic indicators resulting from dams than by
upwards trend, leading to higher flow in these areas. This precipitation in both catchments. This is because the
partly agrees with Souvignet et al., (2014), who found that alteration in climate related factors are more pronounce in
rainfall has increased in recent decades and has led to an larger scale. While the impoundment as dam results more
increase of about 30%-70% in runoff. We found that, the in a local stress. This enables a clear attribution of
upwards trends are only significant in few gauges in both changes to climate or dams, and hence can be used to
areas. Despite the increasing rainfall trend, extreme investigate spatial and temporal patterns of change caused
events become less severe. The changes in precipitation by both factors.
pattern are expected to increase drought intensity
Possible reason for the disparity of change in the two
(Wanders and Wada, 2015). In our case, less low flow
neighboring catchments
events occur but last in a longer duration.
Since the geographical and meteorological conditions
The higher intensity of change in hydrologic alteration
in both catchments are similar, the reasons for the
in Nong Son can be explained by the higher amount of
different change in hydrologic alteration in Thanh My and
annual rainfall in the upstream, especially in Tra My, one
Nong Son are related to the anthropogenic impacts of the
of the highest rainfall location in Vietnam.
dam constructions, particularly the diversion dam.
Hydrologic alteration caused by dam
Kennard et al. (2010) has found that, the assessment
The major changes in Thanh My occur in magnitude of alteration in hydrological regime by using indicators is
and duration of flow, and in Nong Son in the mean impacted by the discharge record length. There, the
monthly flow. The drought period is most affected by expected period is longer than 10 years, ideally 20 years.
damming as agreed by Yan et al. (2010). The most In our case, only changes in precipitation meet this
opposite changes are stressed in flow in early dry season requirement, while the post-period after dam construction,
in the two below-dam gauges which decrease in Thanh 3 and 8 years in Thanh My and Nong Son respectively, is
My and an increase in Nong Son. These differences are much shorter. Although the impacts of damming on
 st
The 1 International Conference on Water Resources and Coastal Engineering
hydrologic regime in this case is more visible and H. (2016). How many RCM ensemble members provide confidence
in the impact of land-use land cover change? International Journal
significant than of precipitation, the less accurate
of Climatology, 37(4), 2080–2100. http://doi.org/10.1002/joc.4836
estimation of alteration by dams in some indicators is [7] Pohlert, T. (2016). Package „ trend ‟: Non-Parametric Trend Tests
unavoidable and can only be mitigated by the availability and Change-Point Detection, 26.
of longer discharge time series after the dam construction. http://doi.org/10.13140/RG.2.1.2633.4243
[8] Rahmani, V., Hutchinson, S. L., Harrington, J. A., Hutchinson, J.
In conclusion, we presented a combined analysis to M. S., & Anandhi, A. (2015). Analysis of temporal and spatial
understand the actual change of the hydrologic regime distribution and change-points for annual precipitation in Kansas,
due to climatic and anthropogenic impacts. The USA. International Journal of Climatology, 35(13), 3879–3887.
http://doi.org/10.1002/joc.4252
quantitative assessment revealed that, the precipitation [9] Richter, B. D., Baumgartner, J. V, Powell, J., & Braun, D. P.
resulted in an increase of mean monthly flows in both (1996a). A Method for Assessing Hydrologic Alteration within
sub-basins. The damming should be responsible for the Ecosystems. Conservation Biology, 10(4), 1163–1174.
contradictory alteration in river regime and the diversion http://doi.org/10.2307/2387152
[10] Richter, B. D., Baumgartner, J. V, Powell, J., & Braun, D. P.
dam cause different hydrological impacts than ordinary (1996b). A Method for Assessing Hydrologic Alteration within
hydropower dams. We can conclude that the relative Ecosystems. Conservation Biology, 10(4), 1163–1174.
contribution of effect of climatic and anthropogenic http://doi.org/10.2307/2387152
factors is distinguishable. This lays a foundation for the [11] Scheffer, M., & Carpenter, S. . (2003). Catastrophic regime shifts in
ecosystems: linking theory to observation. TRENDS in Ecology and
further studies which aim to examine the impacts of more Evolution, 18(12), 648–656.
factors related to climate and anthropogenic impacts on http://doi.org/10.1016/j.tree.2003.09.002
the changes in river regime. [12] Souvignet, M., Laux, P., Freer, J., Cloke, H., Thinh, D. Q., Thuc,
T., … Ribbe, L. (2014). Recent climatic trends and linkages to river
discharge in Central Vietnam. Hydrological Processes, 28(4),
References 1587–1601. http://doi.org/10.1002/hyp.9693
[13] Vu, T. T., Kiesel, J., Guse, B., & Fohrer, N. (2018). Towards an
[1] Costigan, K. H., & Daniels, M. D. (2012). Damming the prairie:
improved understanding of hydrological change - Linking
Human alteration of Great Plains river regimes. Journal of
hydrologic metrics and multiple change point tests. Journal of
Hydrology, 444–445, 90–99.
Water and Climate Change, In press.
http://doi.org/10.1016/j.jhydrol.2012.04.008
http://doi.org/10.2166/wcc.2018.068
[2] Do Hoai Nam Akira Mano, K. U. (2012). Climate Change Impacts
[14] Wanders, N., & Wada, Y. (2015). Human and climate impacts on
on Runoff Regimes at a River Basin Scale in Central Vietnam.
the 21st century hydrological drought. Journal of Hydrology, 526,
Terrestrial, Atmospheric and Oceanic Sciences, 23(5), 541–551.
208–220. http://doi.org/10.1016/j.jhydrol.2014.10.047
http://doi.org/10.3319/TAO.2012.05.03.03(WMH)1.
[15] Yan, Y., Yang, Z., Liu, Q., & Sun, T. (2010). Assessing effects of
[3] Gao, Y., Vogel, R. M., Kroll, C. N., Poff, N. L., & Olden, J. D.
dam operation on flow regimes in the lower Yellow River.
(2009). Development of representative indicators of hydrologic
Procedia Environmental Sciences, 2(5), 507–516.
alteration. Journal of Hydrology, 374(1–2), 136–147.
http://doi.org/10.1016/j.proenv.2010.10.055
http://doi.org/10.1016/j.jhydrol.2009.06.009
[16] Yang et al. (2008). A spatial assessment of hydrologic alteration
[4] Henriksen, J. A., Kennen, J, G., & Nieswand, S. (2006). Users‟
caused by dam construction in the middle and lower Yellow River,
manual for the Hydroecological Integrity Assessment Process
China. Hydrological Processes, 22(March 2008), 3829–3843.
software (including the New Jersey Assessment Tools): U.S.
http://doi.org/10.1002/hyp
Geological Survey. Biological Resources Discipline, Open File
[17] Ye, X., Zhang, Q., Liu, J., Li, X., & Xu, C. Y. (2013).
Report 2006-1093, 71p.
Distinguishing the relative impacts of climate change and human
[5] Kennard, M. J., Mackay, S. J., Pusey, B. J., Olden, J. D., & Marsh,
activities on variation of streamflow in the poyang lake catchment,
N. (2010). Quantifying uncertainty in estimation of hydrologic
china. Journal of Hydrology, 494, 83–95.
metrics for ecohydrological studies. River Research and
http://doi.org/10.1016/j.jhydrol.2013.04.036
Applications, 26, 137–156. http://doi.org/10.1002/rra.1249
[6] Laux, P., Nguyen, P. N. B., Cullmann, J., Van, T. P., & Kunstmann,

(Received on: …/…/2019, Reviewed on: …/…/2019))

 T.T.Vu, J.Kiesel, B.Guse, N.Fohrer 6

Author information
Tinh Thi Vu
PhD student
Faculty of Water Resource Engineering, The University of Danang - University of Science and Technology, 54
Author picture 1
Nguyen Luong Bang, Danang, Vietnam
Topics: Catchment hydrology, ecohydrology, water resources management
Phone: +49 431 880 1268
Dr. Jens Kiesel
Department of Ecosystem Research, Leibniz-Institute of Freshwater Ecology and Inland Fisheries (IGB),
Justus-von-Liebig-Str. 7, 12489 Berlin, Germany
Author picture 2
Topics: Hydrologic and hydraulic modelling; habitat modelling; sediment transport; GIS applications and
Python programming; integrated water resources management
Phone: +49 431 880 1224
Dr. Björn Guse
GFZ German Research Centre for Geosciences, Section Hydrology, 14473 Potsdam, Germany
Author picture 3
Topics: Hydrological and Eco-hydrological modelling; spatial and temporal analysis of flood patterns
Phone: +49 431 880 1224
Prof.Dr. Nicola Fohrer
Department of Hydrology and Water Resources Management, Christian-Albrechts-University Kiel, 24118 Kiel,
Germany
Author picture 4
Topics: GIS-based modeling of water and mass balance on the micro- and mesoscale; effect of land use change;
interdisciplinary modeling of landscape functions; aquatic ecological
Phone: +49 431 880 1276